---
name: scale-data-inventory
description: Documents who owns each data source, where it lives, how it's collected, and how good it is, then flags what will break an AI deployment. Use when the user says "data inventory", "who owns our data", "is our data ready for AI", "where does our data live", "data quality check", or after scale-opportunity-matrix.
---

# scale-data-inventory

Produce `artifacts/scale-data-inventory.md` and update `context/stack.md`. This is the step people skip and regret.

## Preflight (every run)
1. Confirm `GTM-AI-OS/` is accessible. If not: "I need access to your GTM-AI-OS folder. Select it in Cowork's folder picker, or run `os-install` from the core plugin." Stop until it is.
2. Read `CLAUDE.md`, `INDEX.md`, and the last 20 lines of `signals/log.md`.
3. Check the prerequisites below. For anything missing, ask the cold-start questions and offer: (a) run the earlier skill now, or (b) fast path: answer the questions and this skill writes a minimal version of the missing artifact marked `status: draft (fast path)`.
4. On finish, update the artifact row in `INDEX.md` and append one line to `signals/log.md`: `<date> | <skill> | <produced> | <human changed> | <open question>`.

## Prerequisites
- `context/stack.md` at draft or better (from os-install).
Cold-start: "Name your CRM, your call recorder, and where email and enrichment data live." Write to `stack.md` and continue.

## Step 1: Sources
For every system in the friction map's workflow: what data it holds, who owns it (RevOps, IT, GTM engineering, enablement, a vendor), where it flows next, and whether it's reachable by API or MCP.

## Step 2: Collection
For each key field: is it entered by a human, synced, or generated? Apply the 5% rule: of what humans used to enter, about 5% still needs a human. List which fields those are and why.

## Step 3: Quality
Score key fields on complete, accurate, accessible, timely (1 to 5 each). If `~~crm` is connected, offer to sample field completion on the last 200 closed deals.

## Step 4: Blockers
Flag: no named owner; data trapped in a tool with no export; conflicting records across two systems; PII or regulated data with no policy (feeds Module 5 governance).

## Step 5: Write
Write the inventory. Set `context/stack.md` to `status: verified` if the user confirms.

## Hand off
Say: "Now pull the evidence for your ICP. Run `os-win-loss-extract`."